A single measurement of fecal hemoglobin concentration outperforms polygenic risk score in colorectal cancer risk assessment

Abstract

AbstractPURPOSEPolygenic risk scores (PRS) have been proposed and are increasingly used for risk assessment in colorectal cancer (CRC) screening. Fecal immunochemical tests (FITs) are widely recommended and used as dichotomous tests for annual or biennial CRC screening, ignoring the quantitative information on fecal hemoglobin concentration above or below the positivity threshold.MATERIALS AND METHODSWe aimed to assess and compare the ability of a single quantitative FIT and PRS to predict presence of advanced colorectal neoplasms (preclinical CRC or advanced adenoma) as a key indicator of CRC risk. A quantitative FIT (FOB Gold, Sentinel Diagnostics) was employed and a weighted PRS based on 140 common risk variants was determined among participants of screening colonoscopy in Germany. We compared areas under the curves (AUCs) of FIT and PRS for predicting presence of advanced colorectal neoplasm in the entire study population, and in subgroups defined by age, sex, family history of CRC, and history of colonoscopy.RESULTSA total of 3,025 participants aged 50-79 years were included, thereof 523 with advanced colorectal neoplasm and 2,502 participants without neoplasm. FIT clearly outperformed PRS in predicting presence of any advanced neoplasm in the entire study population (AUC 0.721, 95%CI 0.693-0.749 versus 0.591, 95%CI 0.564-0.617, p<0.0001), in younger (50-59 years) and older (60-79 years) participants, both sexes, those with and without a family history of CRC, and those with and without a previous colonoscopy (differences in AUC between 0.110 and 0.186, p=0.03 for those with previous colonoscopy and ≤0.005 for all other subgroups).CONCLUSIONA single quantitative FIT, a low cost, easy-to-administer and universally available test, is more informative for CRC risk assessment than so far established PRS, irrespective of age, sex, family history, or history of colonoscopy.